Date of Award

January 2023

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Michael J. Medvecky


Background: Bone bruises are commonly seen on magnetic resonance imaging (MRI) in acute anterior cruciate ligament (ACL) injuries and can provide insight into the underlying mechanism of injury. There are limited reports that have compared the bone bruise patterns between contact and non-contact mechanisms of ACL injury.

Purpose: (1) To examine and compare the number and location of bone bruises in contact and non-contact ACL injuries. (2) To test the reliability of our proposed bone bruise mapping scheme.

Study Design: Case-Control Study, Level of Evidence, 3.

Methods: Three hundred twenty patients who underwent ACL reconstruction surgery from 2015 to 2021 were identified. Inclusion criteria were clear documentation of the mechanism of injury and magnetic resonance imaging (MRI) within 30 days of the injury on a 3T scanner. Patients with concomitant fractures, injuries to the posterolateral corner, posterior cruciate ligament, and/or previous ipsilateral knee injury were excluded. Patients were stratified into two cohorts based on a contact or non-contact mechanism. Preoperative MRIs were retrospectively reviewed by two musculoskeletal radiologists for bone bruises. The number and location of the bone bruises were recorded in the coronal and sagittal plane using fat-suppressed T2-weighted images and a standardized mapping technique. Lateral and medial meniscal tears were recorded from the operative notes, while medial collateral ligament (MCL) injuries were graded on MRI.

Results: A total of 220 patients were included, with 142 (59.1%) non-contact injuries and 78 (40.9%) contact injuries. Significantly more males sustained a contact injury compared to a non-contact injury (69.2% vs 54.2%, P=0.030), while age and BMI were comparable between the two cohorts. The inter-reader reliability for the location of bone bruises demonstrated good agreement (ICC= 0.882). The bivariate analysis demonstrated a significantly higher rate of combined lateral tibiofemoral (lateral femoral condyle [LFC] + lateral tibial plateau [LTP]) bone bruises (82.1% vs 48.6%, P < 0.001) and a lower rate of combined medial tibiofemoral (medial femoral condyle [MFC] + medial tibial plateau [MTP]) bone bruises (39.7% vs 66.2%, P<0.001) in contact injuries. Similarly, non-contact injuries had a significantly higher rate of centrally located MFC bone bruises (80.3% vs 61.5%, P = 0.003) and posteriorly located MTP bruises (66.2% vs 52.6%, P = 0.047). When controlling for age and sex, the multivariate logistical regression model demonstrated that contact injuries were more likely to have LTP bone bruises (OR = 4.721 [95% CI, 1.147 – 19.433], P = 0.032) and less likely to have combined medial tibiofemoral [MFC + MTP] bone bruises (OR = 0.331 [95% CI, 0.144 - 0.762], P = 0.009) compared to non-contact injuries.

Conclusion: Significantly different bone bruise patterns were observed on MRI based on ACL injury mechanism, with contact and non-contact injuries demonstrating characteristic findings in the lateral tibiofemoral and medial tibiofemoral compartments, respectively. The proposed bone bruise mapping scheme demonstrate good inter-reader reliability.


This thesis is restricted to Yale network users only. It will be made publicly available on 05/22/2026